How my ideal international open data survey would work
It would have a clear easy to find online timetable including key dates
It would be spaced out to avoid clashes with similar surveys
Data used for one survey would be reused by others, that is would not ask exactly the same questions — see point about online form
All the questions would be visible before any attempt is made to answer them.
Any marking system would be transparent and the template with any formula would be online
Any survey would be piloted with users
All questions would be unambiguous
There would be some stability with the core questions over subsequent years so that some real benchmarking is possible. Not an ever moving field of goalposts.
The answers would be quantifiable, so not — how many hack events are there: few, some, many? But instead 4 or less; 5–9; 10 or above
The answers would not be subjective, so not — has open data a low, medium or high impact? But instead…..well what would be a good metric?
It would only have one part
Content would be submitted into a stable online form
The form would be pre-populated with the previous year’s responses so that relevant text could be re-used or overwritten.
There would an open collective discussion around the submission to ensure that results were accurate, realistic and agreed by governments, civil society and businesses etc.
The final submitted answers would be published online
The results would be reproducible
One output would be a clear gap analysis to assist self improvement
Another output would be a two page high level summary of the key findings